target: track explicit test cache counts#11948
Open
sluongng wants to merge 2 commits into
Open
Conversation
bduffany
reviewed
May 15, 2026
| TargetType int32 | ||
| TestSize int32 | ||
| Status int32 | ||
| Cached bool |
Member
There was a problem hiding this comment.
want to mark this as // Deprecated? (I guess we probably can stop writing this value eventually since it can be derived from the other columns?)
| // The number of cached test attempts served from the remote cache. | ||
| int32 cached_remotely_count = 9; | ||
|
|
||
| // The total number of recorded test attempts for this target invocation. |
Member
There was a problem hiding this comment.
Does this include cached attempts? (maybe clarify in the comment)
Comment on lines
+58
to
+61
| // Whether all recorded test attempts for this target invocation were served | ||
| // from cache. Older ClickHouse rows may only populate this legacy field | ||
| // without the explicit cache counts below. | ||
| bool cached = 6; |
Member
There was a problem hiding this comment.
without the explicit cache counts
Is this equivalent to checking total_run_count == 0? (maybe be more explicit in this comment)
|
|
||
| func (t *target) effectiveTotalRunCount() int32 { | ||
| count := t.totalRunCount | ||
| if t.observedResultCount > count { |
Member
There was a problem hiding this comment.
do we expect this to ever happen? if not, maybe add a debug log here (with CtxDebug)?
sluongng
force-pushed
the
sluongng/target-cache-counts
branch
3 times, most recently
from
b7faa3c to
6535b95
Compare
A user reported that the test grid could show cached tests as non-cached. That happened because test target history inferred cache state from timestamps and flattened multiple test attempts down to a lossy bool. This change keeps target history test-only, adds explicit local, remote, and total cached attempt counts to ClickHouse-backed test history, and updates the TAP grid to derive full versus partial cache state from those fields. The primary DB path stays on the legacy bool, and cached remains in TargetStatus as a fallback for older ClickHouse rows that do not have the new count columns yet. This requires an additive ClickHouse migration for TestTargetStatuses. Older rows still render via the legacy bool and timestamp heuristic until they are rewritten, while new rows carry the full cache-attempt semantics.
Address review feedback about ambiguity around when the legacy cached bool should be trusted because older rows do not have explicit count columns. Spell out that total_run_count == 0 is the legacy no-counts signal, mark the stored cached column as deprecated for new rows, and log when observed TestResult counts override summary counts.
sluongng
force-pushed
the
sluongng/target-cache-counts
branch
from
6535b95 to
cc5c124
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
A user reported that the test grid could show cached tests as
non-cached. That happened because test target history inferred cache
state from timestamps and flattened multiple test attempts down to a
lossy bool.
This change keeps target history test-only, adds explicit local,
remote, and total cached attempt counts to ClickHouse-backed test
history, and updates the TAP grid to derive full versus partial
cache state from those fields. The primary DB path stays on the legacy
bool, and cached remains in TargetStatus as a fallback for older
ClickHouse rows that do not have the new count columns yet.
This requires an additive ClickHouse migration for TestTargetStatuses.
Older rows still render via the legacy bool and timestamp heuristic
until they are rewritten, while new rows carry the full cache-attempt
semantics.